Once again about online learning

It would seem that everyone who could have written about DL (distance learning) over the past year. In Russian (and not only) schools and universities, the use of DL, which fell like snow on its head in 2020, basically boiled down to the transfer of offline classes to Zoom or Teams and the acceleration of the production of a variety of educational content. At the same time, a discussion arose about whether the DL should be cheaper or more expensive than full-time. Without repeating what has already been written many times by other authors, I will express my point of view and share my experience in preparing materials for university online education, focusing on methods that almost no one uses.



I'll start with why, in my opinion, the cost of pre-school education cannot be more expensive than full-time education. For a very simple reason: modern offline education these days, by definition, contains an online component. At least, this applies to universities and programs of additional professional education (additional vocational education), perhaps, with the exception of some exotic cases. Those.:

 =   +  
      
      

        
        
        
      

    
        
        
        
      
      

        
        
        
      

    
    

Accordingly, the offline price is made up of these two terms. And if face-to-face classes are turned off (as in 2020), then the cost price can only decrease, because online teaching + proctoring costs are usually much cheaper than organizing face-to-face classes. But on one condition: when the training content is prepared in advance. And if not? I am convinced that almost always this means only one thing: the original course did not initially meet modern requirements (see the main thesis about the online component).



What happened throughout the country (and what about the country - around the world) when universities suddenly switched to online education? Teachers began to feverishly dictate their lectures live in Zoom and YouTube, or pre-recording them, and then uploading them to the same YouTube. Dozens and hundreds, if not thousands of identical lessons in mathematical analysis, etc. general subjects! It turned out laboriously, long and, most often, badly. And in most cases it is absolutely pointless, because these courses have already been prepared dozens of times (and not alone, but by large professional teams), recorded in good quality and posted on platforms like Coursera, Udemy, Stepik, etc.



What's the point in all this? Why was it not possible to select a suitable existing online course like MEP in advance and take it as a basis? Many, by the way, did just that - and in the end they won, tk. made distance learning asynchronous (students first listen to a high-quality online course, and then discuss it at seminars in Zuma). It is clear that this is not always possible. Mainly, due to the fact that the existing MEP courses do not always correspond to a specific program of the university or DPO. Some are more complicated, some are simpler, and some are too big.



Obviously, what is the root of the problem: all modern MEPs are linear and not designed for transformation. Of these, it is very difficult or even impossible to "cut" the necessary fragments for a specific program (and we left out the question of copyright and the legality of the use of third-party MEPs in the educational process). In other words: the existing MEP format is not very suitable for transfer (without compromising the quality of education) from university to university.



Thus, the problem is outlined, and I can now share my experience of solving it. Actually, this is what I have been doing for the past few years, which is reflected in the Nerepetitor website and several articles, for example, here, on Habré. In a nutshell, I have quite a lot of my own teaching materials (videos, articles, calculations, mainly mathematical and engineering), which I gradually organized into a knowledge base consisting of atomic content. As a rule, these are links to my resources marked with metadata, from which it is already easy to collect courses. In order to design the desired course, you just need to write the desired trajectory, and the system will automatically add links to the corresponding media resources.



In order not to look like a theoretician, in conclusion I will give an example of practical application from real life. In 2020 I taught this coursein higher mathematics for HSE humanities students, first in person, and then, starting in mid-November, remotely. With the help of my content base, even at the stage of developing and coordinating the discipline program (for myself), I prescribed an approximate learning path, which, during the course, slightly transformed. Some blocks fell out of it, others were added, and some I had to recreate, but there were very few of them. The final content of the course (including additional closed materials: recordings of lectures, seminars, examples of program code and student work) I posted in the internal LMS of the university. I did it on my own initiative, it did not require serious additional labor costs. A framework course for HSE, compiled from my content, can be viewed here, and to test the generation of different dynamic courses in mathematics by a search query - on the main site of the Nerepetitor project .



So far, the described approach is almost never applied. It's hard for me to say why this is happening. Training providers (platforms, universities and EduTech companies) still prefer to churn out linear MEPs, high-quality, but awkward and with a short life cycle. Perhaps this is due to market demands. But it is important to note: these requests are beginning to change. Looking at related areas, such as CAD systems, we see a growing interest in "generative design" technologies.when a user formulates a short request, and the system automatically (and instantly!) suggests a prototype. Perhaps the same changes in EduTech are not far off too.